Affiliation:
1. Sandia National Laboratories, USA
Abstract
Efficient design of hardware and software for large-scale parallel execution requires detailed understanding of the interactions between the application, computer, and network. The authors have developed a macro-scale simulator (SST/macro) that permits the coarse-grained study of distributed-memory applications. In the presented work, applications using the Message Passing Interface (MPI) are simulated; however, the simulator is designed to allow inclusion of other programming models. The simulator is driven from either a trace file or a skeleton application. Trace files can be either a standard format (Open Trace Format) or a more detailed custom format (DUMPI). The simulator architecture is modular, allowing it to easily be extended with additional network models, trace file formats, and more detailed processor models. This paper describes the design of the simulator, provides performance results, and presents studies showing how application performance is affected by machine characteristics.
Subject
Computer Networks and Communications,Hardware and Architecture
Reference25 articles.
1. M5Sim. (n.d.). The M5 Simulator System: A modular platform for computer system architecture research. Retrieved September 9, 2009 from http://www.m5sim.org
2. Compiler-Optimized Simulation of Large-Scale Applications on High Performance Architectures
3. AMG benchmark summary. (n.d.). Retrieved September 11, 2009 from https://asc.llnl.gov/sequoia/benchmarks/AMG_summary_v1.0.pdf
4. Benveniste, C., & Heidelberger, P. (1995). Parallel simulation of the IBM SP2 interconnection network. Paper presented at the 1995 Winter Simulation Conference, New York, NY.
5. The M5 Simulator: Modeling Networked Systems
Cited by
63 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. MUSE: A Runtime Incrementally Reconfigurable Network Adapting to HPC Real-Time Traffic;2024 IEEE International Parallel and Distributed Processing Symposium (IPDPS);2024-05-27
2. Graph Analytics on Jellyfish topology;2024 IEEE International Parallel and Distributed Processing Symposium (IPDPS);2024-05-27
3. Benefits of Optimistic Parallel Discrete Event Simulation for Network-on-Chip Simulation;2023 IEEE/ACM 27th International Symposium on Distributed Simulation and Real Time Applications (DS-RT);2023-10-04
4. VVQ: Virtualizing Virtual Channel for Cost-Efficient Protocol Deadlock Avoidance;2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA);2023-02
5. HammingMesh: A Network Topology for Large-Scale Deep Learning;SC22: International Conference for High Performance Computing, Networking, Storage and Analysis;2022-11